Identification and characterization of a selective human...

DMD # 82487

1

Identification and Characterization of a Selective Human Carbonyl Reductase 1 Substrate

Diane Ramsden, Dustin Smith, Raquel Arenas, Kosea Frederick, and Matthew A. Cerny

Drug Metabolism and Pharmacokinetics, Boehringer Ingelheim Pharmaceuticals, Inc., Ridgefield,

Connecticut (DR, DS, RA, KF, MAC)

This article has not been copyedited and formatted. The final version may differ from this version.DMD Fast Forward. Published on August 1, 2018 as DOI: 10.1124/dmd.118.082487

at ASPE

T Journals on Septem

ber 20, 2020dm

d.aspetjournals.orgD

ownloaded from

http://dmd.aspetjournals.org/

DMD # 82487

2

Running Title: Identification of a CBR1 Selective Substrate

Corresponding Author: Matthew A. Cerny

Pharmacokinetics, Dynamics and Metabolism

Pfizer, Inc.

Eastern Point Rd.

Groton, CT 06340

Telephone: (860) 441-0223

Fax: (860) 686-7629

Email: [email protected]

Number of text pages: 24

Number of Tables: 2

Number of figures: 6

Number of references: 34

Number of words in abstract: 232

Number of words in introduction: 572

Number of words in Discussion: 1194

ABBREVIATIONS

P450, cytochrome P450; CBR1, carbonyl reductase 1; CBR3; human carbonyl reductase 3;

CBR4, human carbonyl reductase 4; RAAS, renin-angiotensin aldosterone system; ACEi,

angiotensin converting enzyme inhibitors; AKR, aldo-keto reductase; ARB, angiotensin receptor

blocker; CLint, intrinsic clearance; HLCyt, human liver cytosol; HLM, human liver micrsomes;

LC/MS/MS, liquid chromatography coupled with tandem mass spectrometry; LCyt, liver

cytosol; MRA, mineralocorticoid receptor antagonist; NADPH, β-nicotinamide adenine

dinucleotide phosphate reduced form; NADH, β-nicotinamide adenine dinucleotide reduced form;

IS, internal standard; SDR, short-chain dehydrogenase/reductase.


at ASPE


ber 20, 2020dm


ownloaded from

mailto:[email protected]


DMD # 82487

3

ABSTRACT

During drug discovery efforts targeting inhibition of cytochrome P450 11B2 (CYP11B2)-

mediated production of aldosterone as a therapeutic approach for the treatment of chronic kidney

disease and hypertension, (S)-6-(5-fluoro-4-(1-hydroxyethyl)pyridin-3-yl)-3,4-dihydro-1,8-

naphthyridine-1(2H)-carboxamide (1) was identified as a potent and selective inhibitor of

CYP11B2. Pre-clinical studies characterized 1 as low clearance in both in vitro test systems and

in vivo in pre-clinical species. Despite low metabolic conversion, an active ketone metabolite (2),

was identified from in vitro metabolite identification studies. Due to the inhibitory activity of 2

against CYP11B2 as well as the potential for it to undergo reductive metabolism back to 1, the

formation and elimination of 2 were characterized and are the focus of this manuscript.

A series of in vitro investigations determined that 1 was slowly oxidized to 2 by P450 2D6, 3A4,

and 3A5, followed by stereoselective reduction back to 1 and not its enantiomer (3). Importantly,

reduction of 2 was mediated by an NADPH-dependent, cytosolic enzyme. Studies with human

cytosolic fractions from multiple tissues, selective inhibitors, and recombinantly expressed

enzymes indicated that carbonyl reductase 1 (CBR1) is responsible for this transformation in

humans. Carbonyl reduction is emerging as an important pathway for endogenous and

xenobiotic metabolism. With a lack of selective substrates and inhibitors to enable

characterization of the involvement of CBR1, 2 could be a useful probe to assess CBR1 activity

in vitro in both subcellular fractions and in cell-based systems.


at ASPE


ber 20, 2020dm


ownloaded from


DMD # 82487

4

Introduction

The predominant mineralocorticoid, aldosterone, controls salt and water balance via binding to

the mineralocorticoid receptor (MR). Aldosterone actions have also been proposed to occur via

other so-called non-genomic pathways (Funder, 2001; Boldyreff and Wehling, 2003; Vinson and

Coghlan, 2010; Dooley et al., 2012). Though the exact mechanism(s) is not currently understood,

elevated levels of aldosterone are associated with a number of diseases (i.e. chronic kidney

disease, hypertension, and obesity) (Hwang et al., 2013). The renin-angiotensin aldosterone

system (RAAS) has been the focus of much attention by the pharmaceutical industry and has

resulted in multiple drug classes which include direct renin inhibitors, angiotensin converting

enzyme inhibitors (ACEis), angiotensin receptor blockers (ARBs), and MR antagonists (MRAs).

Direct inhibition of the production of aldosterone offers a novel means of treating the diseases

indicated above and may have the additional benefit of decreasing the interaction of aldosterone

through both genomic and non-genomic pathways (Cerny, 2013; Hu et al., 2014; Brem and Gong,

2015; Oparil and Schmieder, 2015).

Drug discovery efforts (Cerny et al., 2015; Weldon et al., 2016) identified 1 ((S)-6-(5-fluoro-4-

(1-hydroxyethyl)pyridin-3-yl)-3,4-dihydro-1,8-naphthyridine-1(2H)-carboxamide, Figure 1) as a

potent inhibitor of aldosterone synthase (CYP11B2). Additionally, 1 only weakly inhibits the

highly similar enzyme, CYP11B1, as well as other steroidogenic and xenobiotic-metabolizing

cytochromes P450. While characterizing the ADME properties of 1, a ketone metabolite (2) was

identified in incubations with liver microsomes and hepatocytes as well as in recombinantly

expressed cytochrome P450 enzymes. Although the abundance of 2 was found to be relatively

low in these systems, additional in vitro characterization of the metabolism of 2 was undertaken

for the following reasons: 1) 2 was shown to have inhibitory properties towards CYP11B2


at ASPE


ber 20, 2020dm


ownloaded from


DMD # 82487

5

similar to that of 1, 2) reductive metabolism of 2 may result in generation of either 1 or its

enantiomer, 3 (Figure 1), therefore, requiring chiral chromatographic analyses of clinical samples,

and 3) reversible metabolism of 2 may contribute to the low clearance observed for 1. Based on

the results of the studies described herein, 2 was determined to be a selective substrate of human

carbonyl reductase 1 (CBR1).

Carbonyl reduction has emerged as an important non-P450 pathway which contributes to the

metabolism of both endogenous substrates and xenobiotics (Malatkova and Wsol, 2014; Cerny,

2016). These reactions can be catalyzed by the aldo-keto reductase (AKR) and carbonyl

reductase family of enzymes and often substrate overlap between these families of enzymes is

observed. NAD(P)(H)-dependent enzymes of both the aldo-keto reductase (AKR) and Short-

chain dehydrogenase/reductase (SDR) families of enzymes have been shown to carry out these

reactions. Carbonyl reductases (CBRs, EC. 1.1.1.184) represent a subfamily of SDR enzymes

that includes three isoforms in humans, CBR1, CBR3, and CBR4. While CBR1 and CBR3 are

both cytosolic enzymes, CBR4 is found in the mitochondria. CBR3 and CBR4 have been shown

to have low carbonyl reducing activity or narrow substrate specificity. In contrast, CBR1 has

been shown to be the most important enzyme of carbonyl reducing enzymes as it is involved in

the metabolism of multiple clinically important drugs, including doxorubicin, daunorubicin,

nabumetone, loxoprofen, haloperidol, pentoxifylline, and bupropion (Rosemond and Walsh,

2004). Despite the recognition that this pathway is important to metabolism of carbonyl

containing compounds, there is still a paucity of data in the literature and further research is

needed. Identification of 2 as a selective substrate for CBR1 may serve as a useful probe for

further study of this enzyme and its involvement in the metabolism of drugs and endogenous

substances.


at ASPE


ber 20, 2020dm


ownloaded from


DMD # 82487

6

Materials and Methods

Reagents β-Nicotinamide adenine dinucleotide phosphate reduced (NADPH), β-nicotinamide

adenine dinucleotide, reduced disodium salt hydrate (NADH), dimethylsulfoxide (DMSO),

acetonitrile (MeCN), phenobarbital, zopolrestat, phenolphthalein, flufenamic acid,

chenodeoxycholic acid, medroxyprogesterone 17-acetate, dexamethasone, indomethacin,

menadione, quercetin ethacrynic acid, dicumarol, 4-methyl pyrazole, disulfiram were purchased

from MilliporeSigma (St. Louis, MO). Compounds 1-5 were from the Boehringer Ingelheim

compound collection and were prepared by published methods (Balestra et al., 2014). Pooled

liver microsomes from mixed gender human (HLM), pooled cynomolgus male monkey (MkLM),

pooled male beagle dog (DLM), and pooled male Wistar Han rat (RLM) were purchased from

Corning (Tewksbury, MA). Pooled mixed gender human,cynomolgus male monkey, male

beagle dog, and male Wistar Han rat liver, kidney, and lung cytosolic fractions were purchased

from Corning (Tewksbury, MA). Pooled human, male cynomolgus monkey, male beagle dog,

and male Wistar Han rat suspension hepatocytes were obtained from Bioreclamation IVT

(Baltimore,MD). Recombinant human carbonyl reductase 1 (CBR1), carbonyl reductase 3

(CBR3), and carbonyl reductase 4 (CBR4) were purchased from Abcam (Cambridge, MA).

Metabolite identification studies in liver microsomes Compound 1 (10 µM) was incubated

with human, monkey, dog, and rat liver microsomes (LM) at a 1 mg/mL final protein

concentration in 0.1 M phosphate buffer (pH 7.4) for 2 h at 37 °C in a total volume of 1 mL

using either NADPH or NADH (1 mM) as cofactor. Additional metabolite identification studies

were carried out on 2 (10 µM) in both HLM and human liver cytosol (HLCyt) as described

above for 1. Incubations were halted by addition of cold MeCN (1 mL) followed by thorough

vortexing and removal of precipitate protein by centrifugation (10 min at 3000xg). Supernatants


at ASPE


ber 20, 2020dm


ownloaded from


DMD # 82487

7

were removed and concentrated to dryness under a stream of nitrogen. The resulting material

was then reconstituted in starting mobile phase and analyzed by LC/MS/MS as described below.

Metabolite identification studies using hepatocytes Compounds 1 (10 µM) was incubated

with hepatocytes (final concentration 2 million cells/mL) in Williams E media for 4 h at 37 °C in

a total volume of 0.5 mL. Incubations were halted by addition of cold MeCN (1 mL) followed

by thorough vortexing and removal of precipitate protein by centrifugation (10 min at 3000xg).

Supernatants were removed and concentrated to dryness under a stream of nitrogen. The

resulting material was then reconstituted in starting mobile phase and analyzed by LC/MS/MS as

described below.

Reaction phenotyping studies Individual recombinant cytochrome P450 isoforms (rCYPs,

50 pmol/mL final concentration) were incubated with 1 (1 µM) in 0.1 M phosphate buffer

(pH 7.4) containing magnesium chloride (3 mM final concentration) at 37 °C. Incubation were

started by addition of NADPH (1 mM final concentration) bringing the total volume to 0.4 mL.

Aliquots (50 μL) were removed at 0, 5, 10, 20, and 30 min and added to cold MeCN (150 μL)

containing internal standard. Quenched samples were submitted to thorough vortexing and

removal of precipitate protein by centrifugation (10 min at 3000xg). Supernatants were removed

and concentrated to dryness under a stream of nitrogen. The resulting material was then

reconstituted in starting mobile phase and analyzed by LC/MS/MS as described below.

Kinetics assessment of ketone reduction Formation of 1 from 2 was assessed in triplicate

incubations in 0.1 M potassium phosphate buffer (pH 7.4) containing varying concentrations of 2

(12 concentrations between 0 – 100 for LCyt and HLM or 0 – 500 μM for CBR1, CBR3, and

CBR4), cytosolic or microsomal fractions (1 mg/mL final protein concentration) from human or

other pre-clinical species or recombinant CBR1, CBR3, or CBR4 (0.01 mg/mL final protein


at ASPE


ber 20, 2020dm


ownloaded from


DMD # 82487

8

concentration), and NADPH (1 mM final concentration) at a final volume of 0.12 mL. For

specified experiments NADPH was replaced with NADH (1 mM final concentration). For all

experiments, incubations were preincubated for 5 min and started by addition of cofactor. After

5 min, the incubations were quenched by addition of MeCN (0.12 mL) containing internal

standard (IS). Precipitated proteins were removed by centrifugation (5 min at 3000xg). Samples

were analyzed by LC/MS/MS as indicated below.

Substrate depletion in cytosolic fractions across species Parent disappearance experiments

were conducted in triplicate (human) or quadruplicate (monkey, dog, or rat) essentially as

described above. For these experiments a single 1 μM concentration of 2 was used as substrate

and incubations were terminated as above at 0, 5, 15, 30, and 60 min. Precipitated proteins were

removed by centrifugation (10 min at 3000xg). Samples were analyzed by LC/MS/MS as

indicated below.

Inhibition of the reduction of 2 to 1 in human liver cytosol The effect of reductase inhibitors

was assessed by incubation of 2 (15 μM) with HLCyt (1 mg/mL), 0.1 M potassium phosphate

buffer (pH 7.4), and NADPH in the presence of 10 or 100 μM of the following inhibitors:

phenobarbital, zopolrestat, phenolphthalein, flufenamic acid, chenodeoxycholic acid,

medroxyprogesterone 17-acetate, dexamethasone, indomethacin, menadione, quercetin,

ethacrynic acid, dicumarol, 4-methyl pyrazole, and disulfiram. Incubations were started by

addition of NADPH and warmed at 37 °C for 5 min. Incubations were quenched by addition of

MeCN containing internal standard. Samples were analyzed by LC/MS/MS as indicated below.

LC/MS/MS Analyses

Metabolite identification Samples (20 µL) were injected on a Supelco Discovery C18 5µ 2.1 x

150 mm column maintained at room temperature using a Thermo Accela High Speed LC system.


at ASPE


ber 20, 2020dm


ownloaded from


DMD # 82487

9

A solvent system consisting of water with 0.1% formic acid (mobile phase A) and acetonitrile

with 0.1% formic acid (mobile phase B) was used at a flow rate of 0.4 mL/min was used with the

following gradient: 0 – 5 min 15% B, 5 – 10 min 15 – 52.5% B, 15 – 16 min 52.5 – 85% B

maintain 85% B for 2 min, 18 – 19 min 15% B then equilibrated for 3 min. The column eluent

was analyzed by MS on a Thermo LTQ OrbiTrap XL Mass Spectrometer run in positive mode

using FTMS and data-dependent FT MS/MS scans (CID and HCD) to elucidate metabolite

structures. An in-line Accela photodiode array detector (Thermo Scientific) obtained

simultaneous UV absorption spectra. The resulting data were analyzed using Xcalibur 3.0 and

Thermo Metworks 1.3 software.

Analysis of kinetics study samples Samples (5 µL) were injected on a Phenomenex Kinetex

C18 1.7 µm, 2.1 x 50 mm column maintained at 40 °C using a Waters Acquity I-Class UPLC

System. A solvent system consisting of water with 0.1% formic acid (mobile phase A) and

acetonitrile with 0.1% formic acid (mobile phase B) was used at a flow rate of 0.5 mL/min was

used with the following gradient: 0 – 0.50 min 10% B, 0.50 – 1.50 min 10 – 90% B, 1.50 – 2.00

min maintain at 90% B, 2.00 – 2.01 90– 10% B, 2.01 – 2.50 reequilibrate at 10% B. The column

eluent was analyzed by MS on an AB Sciex API 6500 Q-Trap Mass Spectrometer run in positive

mode using the following MRM method: for 1 Q1/Q3 317.2 → 254, DP 60, CE 25, for 2 Q1/Q3

315.1 → 252, DP 60, CE 25. A standard curve was constructed by serial dilution in control

tissue sample ranging from 10 – 0.001 M. The resulting MS data were analyzed using Analyst

1.4.2 software.

Chiral chromatography of 1 and 3 Samples (5 µL) were injected on a Chiral-AGP 2.0x100

mm column maintained at room temperature using a Waters Acquity I-Class UPLC System. A

solvent system consisting of water (mobile phase A) and acetonitrile (mobile phase B) was used


at ASPE


ber 20, 2020dm


ownloaded from


DMD # 82487

10

at a flow rate of 0.25 mL/min was used with an isocratic flow of 95% A and 5% B. The column

eluent was analyzed by MS on an AB Sciex API 6500 Q-Trap Mass Spectrometer run in positive

mode using the following MRM method: for both 1 and 3 Q1/Q3 317.2 → 254, DP 60, CE 25.

The resulting MS data were analyzed using Analyst 1.4.2 software.

Calculations

Kinetics assessment of ketone reduction: The rate of formation of 1 from 2 was plotted against

the nominal incubation concentrations of 2 and fit to the Michaelis-Menten equation using

GraphPad Prism 7.02 to determine the kinetics parameters, Km and Vmax. Given the low

lipophilicity of 2, (LogP = 1.7) non-specific binding to microsomes is expected to be low. In

vitro intrinsic clearance (Clint) was calculated from these parameters using Equation 1.

Equation 1 𝐶𝑙𝑖𝑛𝑡 = 𝑉𝑚𝑎𝑥

𝐾𝑚

Substrate depletion in cytosolic fractions across species: The natural log of percent substrate

remaining was plotted versus time in order to determine the rate of depletion (kdeg). From these

data Clint was calculated as described in Equation 2 below.

Equation 2 𝐶𝑙𝑖𝑛𝑡 = 𝑘𝑑𝑒𝑔 x 𝑚𝐿 𝑖𝑛𝑐𝑢𝑏𝑎𝑡𝑖𝑜𝑛

1 mg cytosolic protein

Results

Metabolite Identification

Metabolism studies on 1 were carried out across species in a number of in vitro systems

including LM, hepatocytes, and rCYPs. In incubations of LM with 1 performed in the absence

of NADPH neither consumption of 1, nor formation of metabolites was observed. Generally,

greater consumption of 1 was observed in LM than in hepatocytes across species (Supplemental

Table S1 and Table S2). However, in both human LM and hepatocytes, low turnover (<5 %


at ASPE


ber 20, 2020dm


ownloaded from


DMD # 82487

11

consumed in 60 min, Supplemental Figure S1) was observed for 1. Greater consumption of 1

and formation of metabolites was observed in monkey, dog, and rat LM and hepatocytes than in

human. In samples from these incubations 1 was shown to elute at 10.8 min with prominent

fragmentation occurring through the urea moiety (Figure 2) yielding fragments of m/z 300 and

274.

As shown in Figure 3, three monooxygenated metabolites (m/z 333) of 1 were observed. As the

prominent fragmentation for these metabolites occurs via the urea moiety, similarly to what was

described above for 1, little structural information as to position of metabolism was ascertained

from MS2 data. The identity of one of these three metabolites (tR = 9.2 min) was determined to

be the pyridine N-oxide (4) based on a synthetic standard. The site of metabolism for a second

m/z 333 metabolite (m332_2, tR = 9.8 min) was identified as on the methylene adjacent to the

urea nitrogen. This was established using a deuterated analog of 1 (5) where metabolism of 5

resulted in loss of one of the two deuterium atoms (m/z 334) and could result in formation of the

ring-intact hydroxyl species and/or the ring-opened aldehyde metabolite. The structure of the

third m/z 333 (m332_3, tR = 10.5 min) metabolite was not determined. Subsequent reduction and

oxidation of the aldehyde form of m332_2 gave rise to both the m/z 335 (m334, tR = 8.0 min) and

m/z 349 (m348, tR = 8.5 min) metabolites, respectively, which are only observed in hepatocytes.

The structure of both of these metabolites was confirmed with synthetic standards.

Another minor metabolite with m/z 315 (tR = 12.9 min, -2H) was identified in a number of

matrices. The structure of this metabolite was confirmed with a synthetic standard of 2 which

matched the chromatographic retention and mass spectrometric fragmentation pattern of the

m/z 315 metabolite (data not shown). Follow-up studies determined 2 to be a potent and


at ASPE


ber 20, 2020dm


ownloaded from


DMD # 82487

12

selective inhibitor of CYP11B2 (CYP11B2 IC50 = 9.7 nM, CYP11B1 IC50 = 6200 nM) using

cynomolgus monkey adrenal homogenate assays (Cerny et al., 2015).

Metabolites observed in monkey, dog, and rat LM and hepatocytes corresponded to the

metabolites observed in human matrices. No human unique metabolites were observed, and no

additional metabolites were detected in pre-clinical species LM and hepatocytes. Metabolism of

1 in rCYPs resulted in formation of only oxidative metabolites (4, m332_2, m332_3, and 2) with

CYP2C8, CYP2C9, CYP2C19, CYP2D6, CYP3A4, and CYP3A5 contributing to the

metabolism of 1 (Supplemental Table S3).

Additional metabolite identification studies were carried out on 2 in both HLM and HLCytLCyt.

In these studies, 2 was also found to be relatively stable in HLM and HLCyt incubations

supplemented with NADH (<10% consumed in 60 min). The major metabolite observed in these

studies was reduction of 2 to a metabolite consistent with the enantiomers, 1 or 3, in terms of

mass, retention time, and fragmentation pattern and represents 2.2 and 5.1% of the total UV peak

area, respectively (Supplemental Table S4). When 2 was incubated with HLM and NADPH,

formation of a metabolite consistent with 1 or 3 was also the major biotransformation,

represented ~45% of the UV peak area. A minor monohydroxylated metabolite of 2 with

m/z 331 (0.6% of UV peak area) was also observed under these conditions as well as three other

minor metabolites which were only detectable by MS. Incubation of 2 with HLCyt and NADPH

resulted in significant formation of a metabolite consistent with 1 or 3 which represented almost

99% of the total UV peak area.

Determination of Stereoselectivity of the Reduction of 2


at ASPE


ber 20, 2020dm


ownloaded from


DMD # 82487

13

As reduction of 2 may result in formation of either 1 and/or 3, chiral chromatography was

developed to determine the stereoselectivity of this reduction. Chromatographic resolution of 1

from 3 was achieved as shown in Figure 4A and Figure 4B, respectively. Subsequently, analysis

of a sample of the HLCyt incubation of 2 supplemented with NADPH resulted in a single peak

with a retention time that matched 1 (Figure 4C). Spiking of this sample with 3, followed by

chromatographic analysis resulted in two peaks, an earlier eluting peak corresponding to 3 which

was not present in the incubation sample and the previously observed peak which corresponds to

1 (Figure 4D). This analysis indicated that reductive metabolism of 2 occurs stereoselectively

resulting in conversion of 2 back to 1 with no or minimal formation of 3.

Kinetics Studies

To further characterize the reduction of 2 to 1, kinetic studies were carried out in human

cytosolic fractions from multiple tissues as well as HLM and recombinant CBR1, CBR3 and

CBR4. Initial incubations were run with both NADH and NADPH to determine the optimal

cofactor for supporting this reductive transformation. As the determined solubility of 2 at

physiological pH was ~100 μM, data generated at concentrations exceeding this limit were

excluded from kinetics assessments. Kinetics of reductive metabolism of 1 followed Michalis-

Menten kinetics and generally similar Km values were observed across all matrices (Table 1). In

both HLCyt and HLM, NADPH fortified incubations displayed higher Clint for the reductive

metabolism of 2 (Table 1, Figure 5). Formation of 1 from 2 was observed in NADPH-fortified

incubations of either CBR3 or CBR4 only at high concentrations (Supplemental Figure 2). A

comparison of the NADPH-supported reduction of 2 in cytosolic fractions across human tissues

in terms of Clint resulted in a rank order of activity as follows: intestine > liver > lung ≈ kidney


at ASPE


ber 20, 2020dm


ownloaded from


DMD # 82487

14

(Table 1). Figure 6 shows Clint data determined by disappearance of 2 (1 μM) in NADPH-

fortified cytosolic fractions from liver, intestine, kidney, and lung from human, cynomolgus

monkey, dog, and rat. In general, cynomolgus monkey and dog exhibited similar activity to that

of human across tissues. However, rat displayed significantly less activity across all matrices.

Enzyme Identification Studies

Currently highly selective inhibitors for most carbonyl reducing enzymes are not available.

Therefore, identifying the enzyme(s) responsible for the conversion of 2 to 1 was undertaken

using a set of inhibitors which are known to inhibit one or more carbonyl reducing enzymes.

Inhibition of the conversion of 2 to 1 was determined using HLCyt supplemented with NADPH.

Inhibition as a percent of control was determined at 10 and 100 μM inhibitor concentrations for

the inhibitors listed in Table 3. At the 10 μM inhibitor concentration, <20% inhibition was

observed for all inhibitors except for menadione and quercetin which inhibited 32.6 and 56.4%,

respectively. At the 100 μM inhibitor concentration, menadione, quercetin, ethacrynic acid, and

disulfiram showed 95.7, 81.6, 34.4, and 43.5 % inhibition, respectively. Based on the reported

inhibition of carbonyl reducing enzymes, these results indicate the involvement of a carbonyl

reductase (CBR) or a member of the short-chain dehydrogenase/reductase (SDR).

Discussion

The identification of 1 as a potent and selective inhibitor of aldosterone synthase (CYP11B2)

lead to characterization of its ADME in both in vitro systems and in vivo in pre-clinical species.

In these systems, 1 was characterized as a low clearance compound (< 2.5 mL/min/kg, data not

included). Though 1 was found to be metabolically stable, sufficient turnover of 1 was observed


at ASPE


ber 20, 2020dm


ownloaded from


DMD # 82487

15

to allow for the identification of metabolites, as shown in Figure 2. In rCYPs, human LM, and

human hepatocytes three monooxidation metabolites were observed with m/z 333. One of these

metabolites was confirmed to be the pyridine N-oxide metabolite (4). Another of these

metabolites was determined to be hydroxylated on the methylene adjacent to the urea using the

deuterated analog 5, resulting in the ring-intact hydroxyl species and/or the ring-opened aldehyde

metabolite. Subsequent reduction and oxidation of the aldehyde form of this metabolite likely

gives rise to both the m/z 334 and m/z 348 metabolites, respectively. As formation of these

metabolites likely requires cytosolic alcohol and aldehyde dehydrogenases, respectively, these

metabolites are only observed in incubations of 1 with hepatocytes. The specific position of the

third monooxidation metabolite was not determined. A minor ketone metabolite, 2 (<1% of

parent based on UV in HLM and hepatocytes), was confirmed by synthesis of an authentic

standard. Subsequently, 2 was shown to possess inhibitory activity against CYP11B2 and a high

degree of selectivity relative to CYP11B1. Due to its activity against the pharmacological target

as well as potential for reversible metabolism contributing to the low clearance of 1, studies to

determine enzymes responsible for the formation and subsequent metabolism of this ketone

metabolite were undertaken.

As both NADH and NADPH are capable of supporting reductive metabolism depending on the

enzyme(s) involved (Rosemond and Walsh, 2004), we initially sought to determine the optimal

cofactor and matrix for reduction of 2 to 1 in HLM and HLCyt. In HLMs both oxidative

metabolism and reduction of 2 to 1 were shown to be minor. However, 2 was found to be

metabolized rapidly via reduction in HLCyt supplemented with NADPH. Analysis of samples of

the cytosolic incubation indicated that the reduction of 2 occurred stereoselectively back to 1 and

not to its enantiomer, 3 (Figure 4). Such stereoselectivity for carbonyl reduction is not without


at ASPE


ber 20, 2020dm


ownloaded from


DMD # 82487

16

precedent and has been reported for ketotifen (Breyer-Pfaff and Nill, 2000), nortriptyline

(Breyer-Pfaff and Nill, 2000), and dolasetron (Breyer-Pfaff and Nill, 2004). Furthermore, the

reduction of 2 to 1 in human cytosol fractions across a number of tissues (Supplemental Table S4,

Figure 5) was supported predominantly by NADPH with considerably less reductase activity

observed in the presence of NADH.

The NADPH-catalyzed reduction of 2 to 1 exhibited comparable Km values across tissues and in

recombinant CBR1 (Table 1) though Vmax values varied considerably across tissues. In terms of

Clint, the cytosolic reduction activity was found to increase in rate across human tissues (Table 1)

in the following order: intestine >liver > kidney ≈ lung. However, some caution must be taken

in the determined kinetic parameters (i.e. Km, Vmax, and CLint), as the reported Km values in most

matrices are approaching the solubility limit of 1.

Akin to human, reduction of 2 was also observed in cytosolic fractions from other pre-clinical

species. Similar reductive activities were observed across tissues in human, monkey, and dog.

However, reductive metabolism across tissues from rat was considerably slower than in human

tissues. These observations are consistent with known differences in the function and expression

of CBR1 across pre-clinical species and human. These differences have been described in a

recent publication which evaluated the tissue distribution of CBR1 across pre-clinical species

and reported that 1) cbr1 is not expressed in rat liver, 2) cbr1 demonstrates greater specificity in

steroid reduction, and 3) cbr1 is highly expressed in reproductive tissues (Shi and Di, 2017).

To provide insights into the enzyme(s) responsible for the reduction of 2, a set of inhibition

studies was undertaken. As selective inhibitors for many of the carbonyl reductase enzymes

have not been identified, a set of inhibitors targeting one or more reductase enzymes was


at ASPE


ber 20, 2020dm


ownloaded from


DMD # 82487

17

employed at 10 and 100 μM concentrations. The set of inhibitors is presented in Table 2 and

included: phenobarbital (AKR1A1, AKR1B1), zopolrestat (AKR1B1), phenolphthalein

(AKR1C1, AKR1C2, AKR1C3, AKR1C4), flufenamic acid (AKR1C1, AKR1C2, AKR1C3,

AKR1C4), chenodeoxycholic acid (AKR1C2), medroxyprogesterone 17-acetate (AKR1C1,

AKR1C2, AKR1C4), dexamethasone (AKR1C4), indomethacin (AKR1C1, AKR1C2, AKR1C3,

AKR1C4, CR), menadione (CR/SDR), quercetin (CR/SDR), ethacrynic acid (CR/SDR),

dicumarol (quinone oxidoredutase), 4-methyl pyrazole (alcohol dehydrogenase), and disulfiram

(aldehyde dehydrogenase). As shown in Table 2, significant inhibition was observed for

menadione (96% inhibition), quercetin (82% inhibition), disulfiram (44% inhibition), and

ethacrynic acid (34% inhibition) at a 100 μM inhibitor concentration. Likewise, both quercetin

and menadione also showed significant inhibition at a 10 μM inhibitor concentration. Inhibition

by menadione, quercetin, and ethacrynic acid indicate that a CR/SDR enzyme such as CBR1 is

responsible for the reduction of 2. This is further supported by the location of the reductase in

the cytosolic fraction and the greater rate of metabolism with incubations supplemented with

NADPH rather than NADH. Inhibition of reduction of 2 by the aldehyde dehydrogenase

inhibitor, disulfiram, can be explained by a lack of selectivity for this inhibitor at higher

concentrations. Additionally, disulfiram has been previously shown to inhibit two carbonyl

reductases isolated from rat ovaries (Iwata et al., 1992).

Data obtained from inhibition studies, the greater activity observed in the presence of NADPH

versus NADH, studies with recombinant CBR1, as well the majority of reductase activity being

observed in the cytosolic fraction together indicate that CBR1 is the enzyme responsible for this

transformation in humans and that the reduction reaction is selective for CBR1. The reductive


at ASPE


ber 20, 2020dm


ownloaded from


DMD # 82487

18

activity towards 2 observed across human tissues also appears to be in agreement to the protein

expression profile of CBR1 reported by Hua et al. (Hua et al., 2017).

Though many compounds have been identified as substrates for CBR1, the selectivity of most

towards other carbonyl reducing enzymes and cytochromes P450 are generally poor.

Additionally, many of the probe substrates are small, low molecular weight compounds with

limited utility for MS detection. The selective reduction 2 by CBR1, the relatively slow

oxidation of 1 to 2 by CYP enzymes, the lack of inhibitory potency of 2 against CYP enzymes

(other than CYP11B2), and the favorable MS characteristics for this compound make 2 a useful

probe substrate to assess CBR1 activity in vitro in both subcellular factions and in cell-bases

systems.

In addition to the identification of a potentially useful probe of CBR1, the in vitro

characterization of the metabolic interconversion of 1 and 2 alleviated concerns regarding

inhibitory potency of 2 towards CYP11B2, the stereoselectivity of the reductive metabolism of 2,

and the contribution of reversible metabolism of 2 contributing to the low clearance of 1. In

vitro studies indicated that serendipitously conversion of 2 occurred back to 1. The rate of back

reduction of 2 to 1 is anticipated to result in insignificant circulating levels of 2 contributing to

the pharmacology. Additionally, the minimal formation of 2 in systems lacking CBR1 suggests

that oxidative metabolism of 1 is slow and indicates that reversible metabolism involving 2 is

likely minimal.

Acknowledgements

We thank Xin Guo, Kenneth Meyers, Bryan McKibben, and John Lord for their synthesis of

compounds 1 – 5.


at ASPE


ber 20, 2020dm


ownloaded from


DMD # 82487

19

Authorship Contributions

Participated in research design: Arenas, Frederick and Cerny

Conducted Experiments: Arenas and Frederick

Performed data analysis: Arenas, Frederick, and Cerny

Wrote or contributed to the writing of the manuscript: Smith, Ramsden, Frederick, and Cerny


at ASPE


ber 20, 2020dm


ownloaded from


DMD # 82487

20


at ASPE


ber 20, 2020dm


ownloaded from


DMD # 82487

21

References

Atalla A, Breyer-Pfaff U, and Maser E (2000) Purification and characterization of

oxidoreductases-catalyzing carbonyl reduction of the tobacco-specific nitrosamine 4-

methylnitrosamino-1-(3-pyridyl)-1-butanone (NNK) in human liver cytosol. Xenobiotica

30:755-769.

Atalla A and Maser E (2001) Characterization of enzymes participating in carbonyl reduction of

4-methylnitrosamino-1-(3-pyridyl)-1-butanone (NNK) in human placenta. Chem Biol

Interact 130-132:737-748.

Balestra M, Burke J, Chen Z, Cogan D, Fader L, Guo X, McKibben B, Marshall DR, Nemoto

PA, and Yu H (2014) Patent Number US20140323468A1 Naphthyridines, azaindoles and

related compounds as aldosterone synthase inhibitors and their preparation, pp 185,

Boehringer Ingelheim International GmbH, Germany.

Bauman DR, Rudnick SI, Szewczuk LM, Jin Y, Gopishetty S, and Penning TM (2005)

Development of nonsteroidal anti-inflammatory drug analogs and steroid carboxylates

selective for human aldo-keto reductase isoforms: potential antineoplastic agents that

work independently of cyclooxygenase isozymes. Mol Pharmacol 67:60-68.

Boldyreff B and Wehling M (2003) Non-genomic actions of aldosterone: mechanisms and

consequences in kidney cells. Nephrol Dial Transplant 18:1693-1695.

Brem AS and Gong R (2015) Therapeutic targeting of aldosterone: a novel approach to the

treatment of glomerular disease. Clin Sci (Lond) 128:527-535.

Breyer-Pfaff U and Nill K (2000) High-affinity stereoselective reduction of the enantiomers of

ketotifen and of ketonic nortriptyline metabolites by aldo-keto reductases from human

liver. Biochem Pharmacol 59:249-260.


at ASPE


ber 20, 2020dm


ownloaded from


DMD # 82487

22

Breyer-Pfaff U and Nill K (2004) Carbonyl reduction of naltrexone and dolasetron by

oxidoreductases isolated from human liver cytosol. J Pharm Pharmacol 56:1601-1606.

Cerny MA (2013) Progress towards clinically useful aldosterone synthase inhibitors. Curr Top

Med Chem 13:1385-1401.

Cerny MA (2016) Prevalence of Non-Cytochrome P450-Mediated Metabolism in Food and Drug

Administration-Approved Oral and Intravenous Drugs: 2006-2015. Drug Metab Dispos

44:1246-1252.

Cerny MA, Csengery A, Schmenk J, and Frederick K (2015) Development of CYP11B1 and

CYP11B2 assays utilizing homogenates of adrenal glands: Utility of monkey as a

surrogate for human. J Steroid Biochem Mol Biol 154:197-205.

Deyashiki Y, Taniguchi H, Amano T, Nakayama T, Hara A, and Sawada H (1992) Structural and

functional comparison of two human liver dihydrodiol dehydrogenases associated with 3

alpha-hydroxysteroid dehydrogenase activity. Biochem J 282 ( Pt 3):741-746.

Dooley R, Harvey BJ, and Thomas W (2012) Non-genomic actions of aldosterone: from

receptors and signals to membrane targets. Mol Cell Endocrinol 350:223-234.

Edwards YH, Potter J, and Hopkinson DA (1980) Human FAD-dependent NAD(P)H diaphorase.

Biochem J 187:429-436.

Funder JW (2001) Non-genomic actions of aldosterone: role in hypertension. Curr Opin Nephrol

Hypertens 10:227-230.

Gebel T and Maser E (1992) Characterization of carbonyl reducing activity in continuous cell

lines of human and rodent origin. Biochem Pharmacol 44:2005-2012.

Hara A, Taniguchi H, Nakayama T, and Sawada H (1990) Purification and properties of multiple

forms of dihydrodiol dehydrogenase from human liver. J Biochem 108:250-254.


at ASPE


ber 20, 2020dm


ownloaded from


DMD # 82487

23

Hu Q, Yin L, and Hartmann RW (2014) Aldosterone synthase inhibitors as promising treatments

for mineralocorticoid dependent cardiovascular and renal diseases. J Med Chem 57:5011-

5022.

Hua W, Zhang H, Ryu S, Yang X, and Di L (2017) Human Tissue Distribution of Carbonyl

Reductase 1 Using Proteomic Approach With Liquid Chromatography-Tandem Mass

Spectrometry. J Pharm Sci 106:1405-1411.

Hwang MH, Yoo JK, Luttrell M, Kim HK, Meade TH, English M, Segal MS, and Christou DD

(2013) Mineralocorticoid receptors modulate vascular endothelial function in human

obesity. Clin Sci (Lond) 125:513-520.

Iwata N, Inazu N, and Satoh T (1992) Consecutive treatment of disulfiram inhibits ovarian

carbonyl reductase activity in rats. Jpn J Pharmacol 58:167-173.

Kraemer RJ and Deitrich RA (1968) Isolation and characterization of human liver aldehyde

dehydrogenase. J Biol Chem 243:6402-6408.

Malatkova P, Maser E, and Wsol V (2010) Human carbonyl reductases. Curr Drug Metab

11:639-658.

Malatkova P and Wsol V (2014) Carbonyl reduction pathways in drug metabolism. Drug Metab

Rev 46:96-123.

Matsuura K, Deyashiki Y, Sato K, Ishida N, Miwa G, and Hara A (1997) Identification of amino

acid residues responsible for differences in substrate specificity and inhibitor sensitivity

between two human liver dihydrodiol dehydrogenase isoenzymes by site-directed

mutagenesis. Biochem J 323 ( Pt 1):61-64.

Mylari BL, Larson ER, Beyer TA, Zembrowski WJ, Aldinger CE, Dee MF, Siegel TW, and

Singleton DH (1991) Novel, potent aldose reductase inhibitors: 3,4-dihydro-4-oxo-3-[[5-


at ASPE


ber 20, 2020dm


ownloaded from


DMD # 82487

24

(trifluoromethyl)-2-benzothiazolyl] methyl]-1-phthalazineacetic acid (zopolrestat) and

congeners. J Med Chem 34:108-122.

Oparil S and Schmieder RE (2015) New approaches in the treatment of hypertension. Circ Res

116:1074-1095.

Porter SJ, Somogyi AA, and White JM (2000) Kinetics and inhibition of the formation of 6beta-

naltrexol from naltrexone in human liver cytosol. Br J Clin Pharmacol 50:465-471.

Rosemond MJ and Walsh JS (2004) Human carbonyl reduction pathways and a strategy for their

study in vitro. Drug Metab Rev 36:335-361.

Shi SM and Di L (2017) The role of carbonyl reductase 1 in drug discovery and development.

Expert Opin Drug Metab Toxicol 13:859-870.

Steckelbroeck S, Oyesanmi B, Jin Y, Lee SH, Kloosterboer HJ, and Penning TM (2006)

Tibolone metabolism in human liver is catalyzed by 3alpha/3beta-hydroxysteroid

dehydrogenase activities of the four isoforms of the aldo-keto reductase (AKR)1C

subfamily. J Pharmacol Exp Ther 316:1300-1309.

Vinson GP and Coghlan JP (2010) Expanding view of aldosterone action, with an emphasis on

rapid action. Clin Exp Pharmacol Physiol 37:410-416.

Weldon SM, Cerny MA, Gueneva-Boucheva K, Cogan D, Guo X, Moss N, Parmentier JH,

Richman JR, Reinhart GA, and Brown NF (2016) Selectivity of BI 689648, a Novel,

Highly Selective Aldosterone Synthase Inhibitor: Comparison with FAD286 and LCI699

in Nonhuman Primates. J Pharmacol Exp Ther 359:142-150.

Wermuth B (1981) Purification and properties of an NADPH-dependent carbonyl reductase from

human brain. Relationship to prostaglandin 9-ketoreductase and xenobiotic ketone

reductase. J Biol Chem 256:1206-1213.


at ASPE


ber 20, 2020dm


ownloaded from


DMD # 82487

25


at ASPE


ber 20, 2020dm


ownloaded from


DMD # 82487

26

Figures

Figure 1 Structures of compounds 1-5

Figure 2 Major MS fragments for compound 1 and metabolites.

Figure 3 Metabolic scheme for compound 1.

Figure 4 Chiral chromatography of 3 (A), 1 (B), incubation of 2 with HLCyt + NADPH, and

incubation of 2 with HLCyt spiked with 3.

Figure 5 Comparison of the reductive metabolism of 2 to 1 in human tissues cytosol and

microsomes supplemented with NADPH or NADH.

Figure 6 Comparison of the reductive metabolism of 2 in cytosols across tissues from human,

monkey, dog, and rat supplemented with NADPH.


at ASPE


ber 20, 2020dm


ownloaded from


DMD # 82487

27

Tables

Table 1. Enzyme kinetic parameters for conversion of 2 to 1 in various matrices supplemented

with either NADPH or NADH.

Source Tissue Fraction Cofactor Km Vmax Clint

μM pmol/min/mg μL/min/mg protein

hCBR1 NADPH 107.6 1180 10.97

Human Liver Cytosol NADPH 91.2 15.8 0.17

Cytosol NADH 63.6 0.49 0.0078

Microsomes NADPH 52.3 1.68 0.032

Microsomes NADH 56.4 0.15 0.0027

Intestine Cytosol NADPH 39.8 7.38 0.19

Lung Cytosol NADPH 209 21.4 0.10

Kidney Cytosol NADPH 94.2 9.01 0.10


at ASPE


ber 20, 2020dm


ownloaded from


DMD # 82487

28

Table 2. The effect of reductase inhibitors on the conversion of 2 to 1 in HLCyt supplemented

with NADPH.a

Inhibitor Enzyme Targets % Inhibition

Reference 10 µM 100 µM

Phenobarbital AKR1A1

AKR1B1 2.4 -2.7 (Gebel and Maser, 1992)

Zopolrestat AKR1B1 10.3 15.2 (Mylari et al., 1991)

Phenolphthalein

AKR1C1

AKR1C2

AKR1C3

AKR1C4

6.6 6.0 (Atalla et al., 2000;

Steckelbroeck et al., 2006)

Flufenamic acid

AKR1C1

AKR1C2

AKR1C3

AKR1C4

8.0 18.1 (Matsuura et al., 1997; Atalla

et al., 2000)

Chenodeoxycholic acid AKR1C2 -0.9 -2.9 (Bauman et al., 2005)

Medroxyprogesterone

17-acetate

AKR1C1

AKR1C2

AKR1C4

5.0 6.4 (Hara et al., 1990)

Dexamethasone AKR1C4 0.7 -7.6 (Deyashiki et al., 1992)

Indomethacin

AKR1C1

AKR1C2

AKR1C3

AKR1C4

CBR

-0.2 12.2 (Gebel and Maser, 1992)

Menadione CBR/SDR 32.6 95.7 (Wermuth, 1981; Porter et al.,

2000)

Quercetin CBR/SDR 56.4 81.6 (Wermuth, 1981)

Ethacrynic acid CBR/SDR 11.6 34.4 (Wermuth, 1981; Atalla et al.,

2000; Atalla and Maser, 2001)

Dicumarol Quinone

oxidoredutase 19.7 11.2 (Edwards et al., 1980)


at ASPE


ber 20, 2020dm


ownloaded from


DMD # 82487

29

4-Methylpyrazole Alcohol

dehydrogenase 1.8 2.2 (Atalla and Maser, 2001)

Disulfiram Aldehyde

dehydrogenase 5.9 43.5 (Kraemer and Deitrich, 1968)

a – Incubations were performed at a 15 μM concentration of 2 and a 1 mg/mL protein

concentration of HLCyt for 5 min at 37 °C.


at ASPE


ber 20, 2020dm


ownloaded from


DMD # 82487

30

Figure 1.


at ASPE


ber 20, 2020dm


ownloaded from


DMD # 82487

31

Figure 2.


at ASPE


ber 20, 2020dm


ownloaded from


DMD # 82487

32

Figure 3.


at ASPE


ber 20, 2020dm


ownloaded from


DMD # 82487

33

Figure 4.


at ASPE


ber 20, 2020dm


ownloaded from


DMD # 82487

34

Figure 5.

0 .0 0

0 .0 5

0 .1 0

0 .1 5

0 .2 0

Cl i

nt

(L

/min

/mg

pro

tein

)

L iv e r C y to so l + N A D P H

L iv e r C y to so l + N A D H

L iv e r M ic ro so m e s + N A D P H

L iv e r M ic ro so m e s + N A D H


at ASPE


ber 20, 2020dm


ownloaded from


DMD # 82487

35

Figure 6.

0 .0 0

0 .0 2

0 .0 4

0 .0 6

0 .0 8

0 .1 0

0 .2 0

0 .4 0

0 .6 0

Cl i

nt

(L

/min

/mg

pro

tein

)

H u m an M o n k e y D o g R a t


at ASPE


ber 20, 2020dm


ownloaded from


Identification and characterization of a selective human...

Documents

Transcript of Identification and characterization of a selective human...